一种面向同构集群系统的并行任务节能调度优化方法

节能调度算法设计是高性能计算领域中的一个研究热点.复制调度算法能够减少后继任务等待延时,缩短任务总体调度时间,但是耗费了更多的能量.为此,作者提出一种启发式处理器合并优化方法 PRO.该方法按照任务最早开始时间和最早结束时间查找处理器时间空隙,将轻负载处理器上的任务重新分配到其它处理器上,从而减少使用的处理器数目,降低系统总体能耗.实验结果表明,和已有的复制任务调度算法TDS、EAD和PEBD相比,优化后的调度算法在不增加调度时间的条件下,能够明显减少使用的处理器数和系统总体能耗,从而更好地实现性能和能耗之间的平衡.     

一种新型基于分簇的能量高效数据收集和融合协议

提出一种能量高效的数据收集和数据融合协议(CEDGAP),它基于分簇路由机制,在网络分簇形成之后加入了簇内节点度的控制机制,网络中将会产生一定数量的休眠节点和休眠区域,在数据传输阶段,这些休眠节点不发送数据,它们与处于同一休眠区域的节点在不同轮次间交替被唤醒并往簇头发送数据.文章分析了CEDGAP的影响因素以及协议的时间复杂度.NS-2仿真结果显示,与LEACH相比,CEDGAP提高了网络能耗负载均衡性能,延长了网络生命期.     

基于光森林的弹性光网络能效组播路由频谱分配策略

在弹性光网络中,光树传输组播可以节省链路代价,但较长的光树需要选择更低的调制等级,消耗更多的频谱资源和发射机功耗.提出一种基于遗传算法的光森林组播和光树重配置(GAMF-TR)优化组播的能效路由、调制格式和频谱分配(RMSA)策略.GAMF-TR策略设计染色体编码表示光森林的组播目的节点划分和光路径组合,通过染色体的基因位概率交叉和变异得到更多的光森林RMSA策略,设计了一个频谱分配效率和发射机功耗折中的适应度函数选择能效最高的光森林RMSA策略,并设计在网络资源充足时将组播从光森林重配置到光树传输,进一步减少发射机功耗的消耗.仿真结果表明:提出的GAMF-TR策略获得了组播的最低带宽阻塞率和最高能效性能.     

无线传感器网络中的分组加密算法评测与分析

无线传感器网络节点受到能耗和物理尺寸的限制,选择占用存储少和节能高效的分组加密算法对无线传感器网络非常重要。针对多种分组加密算法,构建评测系统框架,于不同的应用环境和网络结构中,对算法安全性能、存储量和能效指标进行评测。分析评测结果表明:Rijndael加密算法的高安全性、高能效使其最为适合无线传感器网络。     

A Survey of Energy-Efficient Scheduling Mechanisms in Sensor Networks

Sensor networks have a wide range of potential, practical and useful applications. However, there are issues that need to be addressed for efficient operation of sensor network systems in real applications. Energy saving is one critical issue for sensor networks since most sensors are equipped with non-rechargeable batteries that have limited lifetime. To extend the lifetime of a sensor network, one common approach is to dynamically schedule sensors' work/sleep cycles (or duty cycles). Moreover, in cluster-based networks, cluster heads are usually selected in a way that minimizes the total energy consumption and they may rotate among the sensors to balance energy consumption. In general, these energy-efficient scheduling mechanisms (also called topology configuration mechanisms) need to satisfy certain application requirements while saving energy. In this paper, we provide a survey on energy-efficient scheduling mechanisms in sensor networks that have different design requirements than those in traditional wireless networks. We classify these mechanisms based on their design assumptions and design objectives. Different mechanisms may make different assumptions about their sensors including detection model, sensing area, transmission range, failure model, time synchronization, and the ability to obtain location and distance information. They may also have different assumptions about network structure and sensor deployment strategy. Furthermore, while all the mechanisms have a common design objective to maximize network lifetime, they may also have different objectives determined by their target applications. A preliminary was presented in BROADNETS 2006 [29]     

基于传感器网络的图像质量与能耗关系的探讨

无线传感器网络(WSNs)的应用十分广泛。为了拍摄目标物体,将相机安装在各个区域的不同地方。这样分布的相机计算能力受限,不能灵活地处理传感器网络上传输的图像,而这些图像又必须通过传感器网络一跳一跳地传输到中央服务器。为了减少传输消耗的能量,就应压缩图像,但这将降低图像的质量。对有许多相机拍摄相同物体这一情况进行了研究,发现邻点相机的图像可能重叠。通过让中间传感器处理图像并组合重叠部分,可减少传输的总能量消耗,但导致了图像质量的降低。研究不同路由路径图像质量和节能之间的折中机制,对可视化传感器网络的实用性研究很重要。     

无线传感器网络的数据聚合时机分析

数据聚合是无线传感器网络实现节能的一种重要技术。数据聚合的时机直接影响到数据聚合的准确度和延时,是数据聚合的关键问题之一。建立基于泊松过程的数据聚合模型,分析数据聚合时机的概率特征,提出满足一定概率要求和信息数要求的条件下,数据聚合时机的求解方法。仿真表明,理论分析和模拟实验结果基本相符。该结论为数据聚合时机的分析提供新的思路。     

内存云架构的磁盘节能策略

内存云(RAMCloud)的出现改善了在线数据密集型(OLDI)应用的用户体验,但其能耗高于传统的云数据中心。针对该问题,提出一种适用于该架构的磁盘节能策略。首先,引入遗传算法中适应度函数和轮盘赌法,尽量选择更为节能的磁盘进行数据持久化备份;其次,设定合理的服务器内存缓冲区来延长磁盘的平均连续空闲时间,使得部分服务器磁盘在空闲时进入待机状态。仿真实验结果表明,在50台服务器的内存云系统中,该策略能有效节能约12.69%;而缓冲区大小的设定对于节能效果和数据可用性具有双重影响,需权衡考虑。     

An Energy-Efficient Instruction Scheduler Design with Two-Level Shelving and Adaptive Banking

Mainstream processors implement the instruction scheduler using a monolithic CAM-based issue queue （IQ）, which consumes increasingly high energy as its size scales. In particular, its instruction wakeup logic accounts for a major portion of the consumed energy. Our study shows that instructions with 2 non-ready operands （called 2OP instructions） are in small percentage, but tend to spend long latencies in the IQ. They can be effectively shelved in a small RAM-based waiting instruction buffer （WIB） and steered into the IQ at appropriate time. With this two-level shelving ability, half of the CAM tag comparators are eliminated in the IQ, which significantly reduces the energy of wakeup operation. In addition, we propose an adaptive banking scheme to downsize the IQ and reduce the bit-width of tag comparators. Experiments indicate that for an 8-wide issue superscalar or SMT proeessor,the energy consumption of the instruction scheduler can be reduced by 67%. Furthermore, the new design has potentially faster scheduler clock speed while maintaining close IPC to the monolithic scheduler design. Compared with the previous work on eliminating tags through prediction, our design is superior in terms of both energy reduction and SMT support.     

基于WIA-PA传感器网络的能量有效时间同步算法

WIA-PA协议是一种新型工业级无线传感器网络国际标准;当WIA-PA网络内节点数目增多时,传统的参考广播同步算法RBS以及改进型ERBS算法网络开销非常大;针对这一严峻的问题,提出了一种能量有效的局部最优时间同步算法;该算法对相邻的接收节点在多个参考广播消息的条件下求取平均相位偏差,然后对此平均相位偏差进行最大后验估计,最后根据此估计值调整本地时钟;仿真结果表明,所提时间同步算法在同步精度方面比RBS约提高168％,比ERBS约提高24％;同时,同步过程中的网络能耗也大大减少.